by Jonathan and Stian
The excellent Sarah O’Connor has an thought-provoking column about productivity in Tuesday’s FT. One question she poses is whether the reason for Europe’s low productivity is that innovations aren’t diffusing quickly enough through the economy.
The basis for this is research from the OECD that identifies “a growing divide between “frontier” businesses and the rest of the economy”. (The research, by Dan Andrews, Chiara Criscuolo and Peter Gal, is here; this WSJ piece summarises it.) The rationale is as follows. The best businesses in each industry are increasingly productive because they’re using new technologies. But the laggard companies are slow on the uptake, and are much less productive than the leaders. The net result is that overall productivity growth is low.
We wonder whether might be more going on here than just a failure of technological diffusion. Specifically, we suspect that the nature of what the winning businesses are doing is changing in a way that makes it harder for their competitors to catch up.
We know that across the economy, businesses are investing in more intangible assets – things like R&D, software development, organizational development, design, training and marketing. (Indeed, in countries like the US and the UK, businesses are investing more each year in intangibles than in tangible assets – buildings, computers, machines, etc.)
Now, intangible investments have some odd characteristics. Three of these relate to scale, spillovers and synergies.
- Intangibles are often very scalable: once you’ve developed it, the Uber algorithm or the Starbucks brand can be scaled across any number of cities or coffee shops. (Tangible assets like taxis and espresso machines don’t have this advantage: as your business grows, you need more of them.)
- Intangibles have spillovers: sometimes the company that invests in an intangible doesn’t benefit from it. Xerox funded the development of the graphical user interface, but Microsoft and Apple profited from it. What’s more, some firms seem to be systematically good at exploiting the spillovers from other companies’ investments. (A polite way of describing this is “open innovation”.)
- Intangibles sometimes have large synergies with one another. Uber’s software is valuable. Their networks of signed-up drivers in big cities are valuable too. But combined, they are super-valuable: they are what makes Uber a vast cash-machine. Synergies also makes it harder than you might think to copy another company’s intangibles, even if they’re not legally protected. If your competitors assets are synergistic, just copying one element will not yield much return.
When you add these characteristics together, they point to another reason why leading firms might be pulling ahead of their competition. If intangibles are becoming more important to business success, we would see leading companies pulling ahead ways their rivals couldn’t effectively copy:
- They can scale their valuable intangibles across a large business (the OECD makes the point that leading firms tend to have larger sales than laggards). This will make them more productive, since the get a greater return from the same investment. Their smaller rivals don’t get the same benefit even if they make similar investments.
- If they’re unusually good at appropriating spillovers, they can further increase their productivity by copying, exapting or learning from investments made by laggard firms. (Yahoo trains a new product manager, but Google hires her; Nokia designs a new feature for its phone, but Apple quickly introduces something similar.)
- To the extent leading firms have more intangibles than their competitors, it may be hard for laggards to catch up with them, because even where their ideas can be copied cheaply or for free, they may not work as well without the leading firm’s whole suite of synergistic investments. (Copying Uber’s software is not impossible – this fun article estimates the cost of replicating various tech startups’ code; but copying Uber’s software, its driver networks, its political capital and its installed base of users would be insanely expensive.)
All this may seem paradoxical. In an economy that’s more dependent on knowledge, you’d expect it to be easier to compete. After all, knowledge is, as economists say, non-rivalrous – wouldn’t laggard firms be able to copy leaders and increase their productivity?
In fact, the odd characteristics of intangibles – their scalability, spillovers and synergies – may make this more difficult. In an economy with more intangibles, we might expect to see more leading firms benefitting from these types of lock-in, and greater dispersion of productivity of the type the OECD has identified.
So what? Well, it’s important to stress that the intangible interpretation doesn’t mean the OECD’s story is untrue. Diffusion of technology is definitely important , and the things the OECD report calls for, like strong competition policy, access to finance, and good education systems, are still good ideas.
But it does mean that it would be interesting to analyse the relationship between intangible-intensity and the distribution of productivity in the OECD’s data set. Are the industries with big gaps between leaders and laggards also ones where companies make lots of intangible investments?
We don’t have access to the OECD’s data set, but we did make a quick attempt to test this idea using a couple of other data sets: productivity data from the 2013 ESSnet project on Linking of Microdata to Analyse ICT Impact (ESSLait here, report here, data here), and data on the share of total investment that is intangible investment by industry and country from the SPINTAN project, an update of data from www.intan-invest.net. This gives data on the spread of labour productivity from 2000-2010, in, broadly, manufacturing and (market) services for a number of countries, and the proportion of investment in those industries that was intangible.
We ended up with this graph:
It shows the relationship between the change in the productivity spread (the gap in productivity between the best and worst firms), averaged from 2001-07 (we missed out the financial crisis years) and intangible intensity in 2001. Each country is a dot on the graph. So, for example, in manufacturing, Italy and Austria don’t invest very much in intangibles, and have had only a small rise in the manufacturing productivity spread. By contrast, the UK, Sweden and France do invest a lot in intaniglbes, and have had a much larger rise in the productivity spread. The same goes for services.
Together the graphs indicate that productivity spreads rose a lot in countries where industries invest a lot in intangibles. This is consistent with the idea that an industry becoming dominated by intangibles will be one in which the top-most firms will be able to achieve economies of scale and so the top-most firms will be able to break away from the laggards. This is consistent with the frontier firms growing productivity faster than the laggards.
This is a quick-and-dirty analysis. It would be good to test it with the OECD’s bigger dataset. But if the relationship between intangibles and productivity dispersion holds, it suggests that the productivity problem caused by the gap between leading and lagging firms may continue to get worse – because there is something defensible about leading firms’ combinations of intangibles even in relatively competitive markets.
This makes competition policy and access to finance even more important, but potentially less effective in closing the gap – and it means that policymakers may end up running just to keep still. If true, that’s a sobering thought for anyone who cares about the state of the economy.